ellipsis flag icon-blogicon-check icon-comments icon-email icon-error icon-facebook icon-follow-comment icon-googleicon-hamburger icon-imedia-blog icon-imediaicon-instagramicon-left-arrow icon-linked-in icon-linked icon-linkedin icon-multi-page-view icon-person icon-print icon-right-arrow icon-save icon-searchicon-share-arrow icon-single-page-view icon-tag icon-twitter icon-unfollow icon-upload icon-valid icon-video-play icon-views icon-website icon-youtubelogo-imedia-white logo-imedia logo-mediaWhite review-star thumbs_down thumbs_up

5 horrible ad placements that could have been avoided

5 horrible ad placements that could have been avoided Andy Ellenthal

You're a brand advertiser. You know you have to be on the internet, but you're not exactly sure how. So you retain an agency. It comes up with cool creative and formulates a campaign. The buy begins. Everyone's excited. Then all hell breaks loose.

Earlier this month, Business Insider published an article that further exposed the dirty little secret of online advertising: Brands are paying for ads that show up in very inappropriate locations. Maybe inappropriate is the wrong word. I mean really, really bad content -- the kind that, if viewed at work, could get you fired.

There are no analytics that can measure the damage brands can suffer from bad ad placements. Bad placements, or "ad fails," are still an issue that just won't go away. At best, we are talking about wasted impressions and money. Worst case, the advertising has a negative brand impact and affects consumer perception and brand favorability. Just for fun, we did an accidental mini case study at the office.

First, my disclaimer: My company evaluates what is on a web page to help a buyer or a seller target. Some of the more interesting URLs that surface catch the attention, and maybe the imagination, of the employees. Thus, we were introduced to the Big Girls Bras brand. By using this example, we are not picking on anyone. In fact, given the wide variety of URLs we see, we often say "we do not judge." Trust me, this example is tame.

At our marketing director's recommendation, a few others visited this site to see what the brand was all about. Lots of team members had a chuckle, but the site dropped a cookie on our browsers and the ad followed us all around the web. All of the placements that follow were seen by Peer39 employees within a three-day span. The placements ranged from being an uncomfortable association to an outright waste. But the good news is that there is an easy fix. We'll explore solutions to each of the following bad placements in this article.

Don't ask me how we got there, but one of the first places the Big Girls Bras ad appeared was on a blog that discussed Jewish women's issues. This article, as you can see from the image, dealt partly with criticizing women because of their clothes, and it also contains the less than desirable term "slutwalk" in the headline. That's not a term any brand wants to be associated with.

What went wrong: When a cookie is dropped, the user gets placed in a bucket and classified as in-market to buy a brassiere, whether that consumer is female or male. The brand knows we visited its site, so it's very likely to buy impressions for us again. With this site, the ad-buying platform likely saw a page with content that appealed to women and felt it was appropriate context for the ad.

How to prevent it: Whether you're using standard retargeting to get site visitors to return to your site or buying intenders via a real-time bidding (RTB) platform, set your controls to weed out inappropriate terms and language. Women's interest sites are great places to advertise women's clothing, but advertisers have to make sure they get appropriate placements.

In the previous example, the targeting engine's logic was easily discernable. I guess, theoretically, one could argue the same can be said for this placement, which appears next to an article about Andy Dick standing trial for sex abuse charges. Celebrity gossip gets lots of page views, so it's understandable why the brand would buy impressions on this site.

What went wrong: The brand might have been looking to align itself with celebrity gossip, and it's also possible that it was going after a keyword like "sexy." That strategy backfired, and it got a placement next to a story about sex abuse. This page is affiliated with a top-tier TV network, which shows that even if you limit your buy to premium sites, there is cause for concern.

How to prevent it: A lot of media buyers new to RTB will often simply make white and black lists of sites they want and don't want. The fact that this ad appeared on a premium publisher site speaks to the problems with this tactic. Instead of making site lists, focus on content you don't want your brand aligned with. There's an obvious difference between sexy and sex abuse, but if you set your RTB controls to avoid negative content, you'll preemptively avoid adjacencies that can jeopardize your brand.

This placement is slightly discouraging. The brand does denote its product is made for big girls, but this placement seems to indicate that the brand assumes its target audience likes to eat. It's not as racy as our first two examples, but it's a potentially uncomfortable brand association. It's akin to seeing a Listerine "fight bad breath" ad on Facebook right next to your picture. What's more likely in this scenario: The consumer clicks the ad, or they take a hard look in the mirror?

What went wrong: A bad targeting decision. It appears that the brand decided food pages were within its desired page placements, and possibly even went after the keyword "fat." Far be it from me to criticize how a brand goes after its audience, but running ads in this questionably appropriate environment leaves the potential for a sour taste in the consumer's mouth and makes it more difficult to deliver successful online advertising.

How to prevent it: This is a great example of the limitations of keyword, or contextual, targeting when it comes to display ads. By using a different approach based on page meaning, this page would have been recognized as a food page. If this advertiser used a layered approach, it could have also chosen content categories like health or weight loss. Such an approach boosts the assurance that the ad will show up on appropriate pages that resonate with the brand message.

Washington seems to be at war with itself, so everyone's looking for the latest political news. But wait a second -- did that Big Girls Bras ad really follow me here? At this point, there's no logical reason for this placement. I know why it happened, and you probably do too. But consumers are at best confused and at worst down-right creeped out that this ad followed them around the web.

What went wrong: On a macro level, this placement sounds good. It's a premium newspaper site and a piece of content that is this particular paper's bread and butter. We've all heard the term "banner blindness," and one of the key causes is a lack of relevance, like we see here. Consumers aren't going to stop reading an article to engage with an ad if that ad doesn't appeal to their interests or state of mind at the time. That's the case here.

How to fix it: Relevance is everything. I was served this ad because a cookie in my browser told the ad server I visited the Big Girls Bras site. I've talked a lot so far about the problems with keywords and exploring your filter options, but this example shows something even more basic. It's not just about the site level or finding the appropriate consumer. Page context is the piece that makes ads truly effective.

Remember what I was just saying about context? It's slightly humorous to see the phrase "Always On" next to a bra ad, but this doesn't belong here.

What went wrong: Let's start with what went right. Most of the above examples are either premium sites or very specific to the audience the brand is trying to reach. This example shows a brand willing to explore smaller, long-tail niche sites to reach consumers. That's a tactic that can pay huge dividends for display advertisers, who can very well put their ads in front of interested consumers while paying less per impression. The problem, again, is simply bad context.

How to prevent it: At this point, frequency capping might be the best defense. But brand marketers using any kind of targeting -- whether it's an agency trading desk, a demand-side platform, an ad exchange, or a retargeting network -- need to ask their ad partners for environmental targeting. Don't be afraid to buy impressions in the long tail. Just make sure you're factoring in what's on the page.

Andy Ellenthal is CEO of Peer39.

On Twitter? Follow iMedia Connection at @iMediaTweet.

Andy Ellenthal is DG’s Executive Vice President of Sales and Ad Operations for both the TV and DG’s online division, MediaMind. Andy comes to DG with extensive management experience in advertising technology companies, including key...

View full biography


to leave comments.

Commenter: Ross Bradley

2011, August 31

Just popped back looking for a reply? In regards to BIG Brands and "re-targeting" (and in a 'no holds barred' fashion), across the entire web.- In my own linked post I point out, that in targeting people not pages, brand safety shouldn't be any concern whatsoever. As it always becomes a situation of (only), being between the "user" who is being re-targeted and the Brand's Ad, that is used to re-target with.

Commenter: Aaron Richman

2011, August 22

Great article Andy. This is a recurring question from many of our advertisers and an issue that prevents many brands from testing retargeting in the first place. As a marketer, it's imperative you know where your ads are running, both from a brand safety and ROI standpoint.

With platforms like AdRoll (disclosure: I work there), advertisers can see the websites their ads are running on along with the costs associated with advertising on those sites. This is important for a couple reasons.

1. At AdRoll, we work very closely with our partners (Google, Yahoo!, Microsoft, AOL + 30 other networks) to ensure our inventory is brand-safe. However, as a marketer, you still want to make sure your ads are appearing on sites that showcase your brand.

2. This list can help you determine where your best prospects are going after they leave your website. For instance, you might see that a high percentage of your audience is visiting TechCrunch.com, and therefore this could be a good channel for lead generation.

3. If you're serving high impressions on a site and it's not bringing you clicks and conversions (e.g. you sell baby products and you see a high volume of ads being served on Wired.com), you can easily remove this site from your campaign.

4. ROI: We show our advertisers what it cost them (CPC & CPM) to run on each website. If our advertisers feel the price is too high, they can exclude the site from their retargeting network.

Hope this helps!

Aaron R.
AdRoll [dot] com

Commenter: Ross Bradley

2011, August 18

Interesting, but? ........."a few others visited this site to see what the brand was all about. Lots of team members had a chuckle, but the site dropped a cookie on our browsers and the ad followed us all around the web."

Did you expect anything other than this possibility? I have only just posted on this very subject and would welcome you to correct me as to where I have gone wrong with my own impression - in regards to BIG Brands and "re-targeting" (and in a 'no holds barred' fashion), across the entire web. I look forward to your reply here. Thanks.


Commenter: Spencer Broome

2011, August 18

Good points and entertaining. It's hard not to find humor in some of these placements. But it's much more humbling when it is your own ad.

Commenter: Ben Plomion

2011, August 18

Great (and also very entertaining) article Andy. Brands need to have the confidence that their ads will only appear on brand-safe sites. There are multiple companies out there like Peer39 that helps rank publishers. Chango has partnered with adSafe and DoubleVerify to ensure that we only buy ads across sites that are brand-safe. Frequency cap should also be used to avoid the big brother effect.